Citation

Ramanathan, Thirumalaimuthu Thirumalaiappan (2023) Hybrid fuzzy method based secured multi-agent system for medical diagnosis. PhD thesis, Multimedia University.

Full text not available from this repository.

Abstract

Medical data mining is a complicated problem that needs efficient artificial intelligence (AI) techniques to overcome various research challenges as medical datasets contain uncertain and incomplete data. Fuzzy logic method can support this issue as it can deal with uncertain data but it also has some drawbacks. A fuzzy expert system can lead to wrong decision making if the fuzzy sets and rule base is not appropriate. The curse of dimensionality is one of the major problems in implementing the fuzzy expert system. To address these issues, this thesis proposes an enhanced parallel hybrid fuzzy reasoning method called Machine Learning based Parallel Fuzzy Reasoning (ML-PFR) method which integrates parallel fuzzy systems with machine learning classifier for the classification and prediction of medical data. To support the classification process, this thesis proposes an enhanced feature selection method based on random forest and ensemble machine learning classification. The ML-PFR method is implemented through a multi-agent approach in this research to work effectively for medical diagnosis problem. Multi-agent system (MAS) is a part of distributed AI concept that will consist of many autonomous problem-solving agents that shares same data and do interaction to make a decision. Security is important in the MAS developed for healthcare applications because if the medical data involved with the agents are corrupted, it may lead to wrong decision making. A novel MAS called Blockchain based Multi-Agent Fuzzy Reasoning System (BMAFRS) is proposed in this thesis that uses ML-PFR method for medical data mining and blockchain concept for agent security. In BMAFRS, several parallel hybrid fuzzy reasoning systems which are developed based on the ML-PFR method are used as the classification agents for classification and prediction of medical data.

Item Type:	Thesis (PhD)
Additional Information:	Call No.: R859.7.F89 T45 2023
Uncontrolled Keywords:	Fuzzy systems in medicine
Subjects:	R Medicine > R Medicine (General) > R858-859.7 Computer applications to medicine. Medical informatics
Divisions:	Faculty of Information Science and Technology (FIST)
Depositing User:	Ms Nurul Iqtiani Ahmad
Date Deposited:	29 Aug 2024 02:33
Last Modified:	29 Aug 2024 02:33
URII:	http://shdl.mmu.edu.my/id/eprint/12878

Downloads

Edit (login required)